Attempt to achieve torque control of an induction motor via a tachometer and off the shelf vfd

[johansen]

Hello folks,

I have not found anyone else who has attempted to do what I'm going to try and describe here:

Take an off the shelf vfd programmed for a 0-10 volt input across the intended rpm range of the induction motor

Then take a dc tachometer setup directly driven by the motor so that it generates the same dc volts, as does the vfd require, to command the motor to run at a given speed.

In theory, the VFD will not run the motor, as the slip will be zero, and if the motor is run up to speed by some other means, the vfd will follow, lagging behind due to the time required for the vfd to increase the frequency.

Now what you do is take a +/- 1 volt supply and add it to the tachometer, using a joystick and a potentiometer. Now you are leading or lagging the motor and the acceleration limit is the vfd software and the slip (nominally 1 volt is 6hz to a vfd)

So anyhow to limit the current:

Take a portion of the dc current on the VFD and run it through a solenoid such that it resists the joystick. you now have force feedback...

I have not had time to put this together yet but I have a battery pack capable of 10kw at 100 cells in series (lifepo) and I have a 5hp 1750 rpm motor that I can wire in delta for 138volts ac

I would like to belt drive it 2:1 to my pickup truck engine... going to test the concept out on a fuji 1/2hp vfd i bought and a 90 watt induction motor, I only need to look around for a good choice for a tachometer at this point.

I'm intending to use the igbt block and most of the circuit board off of 4-5 ton heat pumps and just run a ribbon cable from the 1/2hp vfd to the bigger igbt. if the concept works what i'm thinking i would do to get more power out is to have 1 igbt module drive each set of 3 phase windings. the 4-5 ton pumps are nominally 20 amps at 240v so they typically have 30 amp rated igbts.

For efficiency purposes, at and below idle speed on the gas engine the vfd would have a very low volts/hz curve programmed.

 

[Jrbe]

Too many workarounds to make something you have end up working not so well. Also missing a lot of dead end details to be able to successfully cobble a bunch of stuff together that will make this difficult to implement.

Consider looking at a belt assist hybrid setup and see if it makes sense for what you want to do. Hyundai hsg (~300v bldc,) Dodge belt assist hybrid (48v but induction so not a great option unless the controller can run it), and a few other options. There are people trying to hack these controllers. It's been a while since I looked, last I knew, no one had cracked into them.

A big challenge is getting a torque request if your existing setup is not OBDII. Past that there are no plug and play hybrid controllers (or hacked ones) that I'm aware of. You could potentially write code to pull OBDII info and translate that into instructions for the motor controller to run a VESC or similar controller.

There are a bunch of details I left out. This isn't exactly simple.